OpenCandle Guide

OpenCandle is an open-source AI-driven financial analysis agent developed by Kahtaf. It integrates real-time market data, multi-analyst workflows, and portfolio management tools into a terminal interface, acting as an intelligent research assistant for developers and quantitative analysts. Its core advantage lies in simulating the multi-agent collaboration model of a real research team, combining the reasoning capabilities of large language models to comprehensively evaluate investment opportunities from multiple dimensions such as technical, fundamental, macroeconomic, and sentiment analysis.

Project Background

In the field of financial investment, accessing real-time data, conducting in-depth analysis, and making decisions are complex and time-consuming processes. OpenCandle aims to simplify this workflow using AI technology, providing efficient and comprehensive research tools. As an open-source project, it allows developers and quantitative analysts to freely extend its features to meet personalized needs.

Core Features

Real-Time Market Data Access

• Data Source Integration: Supports data across asset classes like stocks (stock price, trading volume, P/E ratio), cryptocurrencies (real-time price, trading volume), and macroeconomic indicators (interest rates, inflation data).
• Data Processing Pipeline: Implements efficient cleaning, standardization, storage, and indexing, supporting millisecond-level analysis responses.

Multi-Analyst Workflow

Adopts a multi-agent architecture to simulate professional team collaboration:

• Role Division: Technical analysis expert (price trends, indicators), fundamental analyst (financial statements, industry position), macro analyst (economic cycles, policies), sentiment analyst (news, social media).
• Collaboration Mechanism: Agents share information, synthesize conclusions, and follow a preset workflow (macro evaluation → fundamental screening → technical analysis → sentiment verification) to generate recommendations.

Portfolio Tools

• Portfolio Tracking: Real-time calculation of total value, profit/loss, allocation ratio, risk indicators (volatility, maximum drawdown), and comparison with benchmark indices.
• Risk Analysis: Based on modern portfolio theory, provides asset correlation matrix, concentration risk identification, and rebalancing suggestions.

Technical Architecture

Terminal-First Design

• Advantages: Efficient keyboard operations, scriptable, low resource consumption, developer-friendly (target users are familiar with command lines).
• Tools: Uses Python libraries like Rich and Textual to beautify terminal output, providing visual elements such as tables and charts.

AI Model Integration

• Supported Models: OpenAI GPT series, open-source models (run locally via Ollama), and dedicated financial models (if available).
• Prompt Engineering: Carefully designed templates to extract key indicators, define analysis tasks, and standardize output formats (JSON, Markdown tables).

Extensible Architecture

• Plugin System: Supports custom data sources, analysis modules, and output formats.
• Configuration-Driven: Defines workflows via configuration files without modifying code.
• API Interface: May provide a REST API to allow external systems to call analysis capabilities.

Application Scenarios and Challenges

Application Scenarios

• Individual Investors: Obtain institutional-level analysis reports, monitor portfolios, discover opportunities, and verify hypotheses.
• Quantitative Researchers: Test analysis factors, backtest historical performance, and generate research reports.
• Developer Learning: Learn about LLM applications in finance, multi-agent design, financial data processing, and API integration.

Limitations

• Data Quality: Free data sources have latency and frequency limitations; professional trading requires paid data.
• Model Hallucination: Large language models may generate incorrect information; key conclusions need verification.
• Regulatory Compliance: Users must ensure their usage complies with local investment advisory service regulations.

Future Directions and Summary

Future Development

1. More Data Sources: Integrate futures, options, foreign exchange, and alternative data (satellite imagery, supply chain data).
2. Visualization Enhancement: Add terminal chart drawing capabilities.
3. Backtesting Framework: Implement strategy backtesting and performance attribution.
4. Community Contribution: Establish mechanisms for sharing analyst roles and templates.
5. Voice Interaction: Support voice commands for conversational analysis.

Summary

OpenCandle is an innovative application of AI in the financial field. By simulating human research teams through multi-agent collaboration, it provides users with a comprehensive intelligent analysis assistant. It will not replace human investors but enhance their decision-making capabilities. For developers, it is an excellent case for learning AI application development, demonstrating practices in LLM integration, multi-agent design, and terminal tool construction.